PhD Student Wins 1 of 12 Microsoft Fellowships
University of Illinois computer science Ph.D. student Chi Wang has received one of 12 Microsoft Research Fellowships for his work in information network analysis. Wang’s recent work in the field has focused on discovering social relationships in information networks.
His paper entitled “Dynamic Social Influence through Time-dependent Factor Graphs” explores the phenomenon of social influence, and addresses the problem of how to quantify the influence between two users on a given social network. Wang proposed a pairwise factor graph to model influence, a novel algorithm to learn the model and make inferences, and a dynamic factor graph to incorporate time information. The three approaches, when combined, are able to facilitate the application of influence among a network.
In another paper, “Learning Relevance from Heterogeneous Social Network and Its Application in Online Targeting,” Wang proposes a new method for modeling user interest in particular content from heterogeneous data sources with distinct but unknown importance. Wang’s model leverages links in a users social graph by integrating the conceptual representation of a user’s linked objects. Wang and his colleagues applied their approach to the task of selecting relevant ads for Facebook users, which led to an overall improvement in clickthrough rate prediction.